NEW YORK – A team from Columbia University's Mailman School of Public Health, the National Institutes of Health, and Texas A&M University has identified a gene expression signature that appears to predict immune responses to Ebola virus infection and may inform future efforts to curb severe, hemorrhagic forms of the disease.
"Since the current Ebola therapeutics being tested in the [Democratic Republic of Congo] are most effective when given as early as possible in infection, our model could be used to develop tests with a huge impact on clinical care and patient outcomes," senior author Angela Rasmussen, a researcher at the Mailman School of Public Health, said in a statement.
As they reported in a paper in Cell Reports on Tuesday, Rasmussen and her colleagues began by screening 10 mouse strains from the so-called Collaborative Cross (CC) inbred mouse panel, developed from eight parental mouse lines, focusing on strains with naturally variable Ebola virus susceptibility.
Because the clinical data collected during a disease outbreak "rarely contains the combined breadth and specificity of information scientists need to perform detailed analyses of immune function," the investigators reasoned that "[m]ouse models can help fill in the information gap," co-first author Atsushi Okumura, a researcher affiliated with Columbia University and the National Institute of Allergy and Infectious Diseases, explained in a statement.
When they sequenced the RNA of liver and spleen samples collected over time from mice with Ebola virus susceptibility or resistance, the researchers saw relatively early antiviral immune activity in mice from the strains that could better withstand the virus. In contrast, they reported, the more susceptible mice were more prone to have dialed down immune gene activity at early stages in the infection process, followed by rampant inflammation.
"Tolerance is associated with tightly regulated induction of immune and inflammatory responses shortly following infection, as well as reduced inflammatory macrophages and increased antigen-presenting cells, B-1 cells, and [gamma delta] T cells," the authors wrote, while lethal cases were marked by "suppressed early gene expression and reduced lymphocytes, followed by uncontrolled inflammatory signaling, leading to death."
Together with clinical features, information on viral loads in the infection-tolerant or -vulnerable mice, microRNA sequence data, machine learning analyses, and other approaches, the RNA sequence data helped the team narrow in on a set of genes with distinct expression profiles in mice that died from Ebola virus disease and those that survived.
When they applied this potential Ebola virus response signature to blood RNA-seq data from Ebola virus patients treated in Guinea between 2014 and 2016, the researchers found that it could accurately predict infection responses roughly three-quarters of the time. Though they cautioned that further work is needed to verify the expression-based ties to Ebola virus outcomes, they noted that the findings so far hint at the potential clinical utility of their mouse modeling approach.
"While this study focused on tissues that drive [Ebola virus disease] pathogenesis, future work should also examine transcriptomic data obtained from CC mouse peripheral blood, which are more comparable to clinically relevant human samples," the authors noted, adding that the disease response marker candidates identified in the current study "must also be validated in other large [Ebola virus disease] transcriptomic datasets, ideally from humans and [non-human primates] with differential outcomes."